Secondary Logo

Journal Logo

Clinical Cardiovascular

Leukocytes-Depleting Filters Preferentially Remove Activated Leukocytes and Reduce the Expression of Surface Adhesion Molecules during the Simulated Extracorporeal Circulation of Human Blood

Alexiou, Christos*; Sheppard, Stuart*; Tang, Augustine*; Rengarajan, Arvind*; Smith, David; Haw, Marcus*; Gibbs, Roz

Author Information
doi: 10.1097/01.mat.0000225894.61493.2f
  • Free


The use of cardiopulmonary bypass (CPB) in cardiac surgery is linked with a whole-body inflammatory reaction.1–4 Because leukocyte activation, during this process, is an important step contributing to end-organ injury, leukocyte removal with special filters has been proposed as a means of reducing the unwanted effects of CPB.4,5

The leukocyte-depleting filters remove leukocytes by a combination of physical and biologic mechanisms, the latter involving the process of leukocyte adhesion,6–8 an active process that is independent of mechanical forces. It has been suggested that adhesion plays a more important role than the physical mechanisms, because the cells trapped inside the filter are much smaller then the pores of the filter.9–11

Because activated leukocytes are more adhesive than their nonactivated counterparts,12,13 it would be reasonable to hypothesize that they would be removed at a higher rate than the nonactivated cells through an active adhesion process. Clinical and experimental studies claim that these filters remove preferentially the activated forms of leukocytes from the circulation.14–19 Other studies, however, describe no differences in the extent of leukocyte activation with regard to the use of leucodepletion.20,21

In vitro experimental studies utilizing a specially designed simulated extracorporeal circuit, a miniature leukocyte-depleting filter, and fresh whole human blood were performed to examine the effect of a leukocyte-depleting filter on the manually counted total white cell counts (WCC), and to ascertain the effect of this filter on the activation status of leukocytes assessed with two approaches: (1) flow cytometric quantification of the expression of the surface adhesion molecules (and markers of leukocyte activation) CD11b, CD18, and CD62L; and (2) manual counting of the numbers of circulating activated leukocytes.

Materials and Methods

The protocol of this study was reviewed and approved by the Ethics Research Committee at Southampton General Hospital and informed consent was obtained from the patients recruited.

A 200 ml blood sample was taken from 10 patients undergoing first-time elective coronary artery bypass surgery. Blood was taken 5 minutes after the start of CPB (to ensure sufficient equilibrium between patient blood and the fluid used to prime the CPB circuit) from the arterial outlet of the oxygenator into a graduated fluid bag. At the same time, an additional 1 ml blood sample was taken and analyzed for blood gas and hemoglobin concentrations.

Inclusion Criteria

Blood samples were taken from patients who had normal preoperative hematology screenings, and a body weight in excess of 60 kg so as to reduce the effects of hemodilution during CPB. Patients receiving steroids or other antiinflammatory medications were not included in the study. Jehovah's Witness patients were also excluded due to their avoidance of receiving blood products.

Specimen Collection

The 200 ml blood sample was circulated within an experimental extracorporeal circuit (described below) for 60 minutes, while maintaining a blood temperature of 36°C. This blood temperature was chosen because previous experimental and clinical work conducted by our group16,22,23 showed that the overall leukodepleting efficacy of the LG6 filter (Pall, Portsmouth, UK) is enhanced during warm CPB. A specially designed leukocyte-depleting filter (LG6, Pall) was attached to the extracorporeal circuit while blood taken from five of the patients was circulated. No filter was used during the circulation of the blood taken from the remaining five patients, who were used as controls. The latter allowed the estimation of the nonspecific absorption of leukocytes by the tubing material of the circuit. During the 60 minutes that the 200 ml of blood was circulated within the extracorporeal circuit, 1000 μl blood samples were taken every 10 minutes to determine the total and activated WCC manually under light macroscopy and to analyze the expression of CD11b, CD18, and CD62L by flow cytometry.

Thus, the leukodepletion (n = 5) and control (n = 5) groups were compared against the following end-points: (a) total WCCs, (b) activated WCCs, and (c) expression of the surface adhesion molecules (and markers of leukocyte activation) CD11b, CD18, and CD62L. In addition, the correlation between the two methods used to quantify the activation of leukocytes (activated WCCs and expression of surface adhesion molecules) was examined.

Construction and Validation of the Extracorporeal Circuit

A miniaturized circuit incorporating a miniature leukocyte-depleting LG6 filter was constructed (Figure 1) in order to recreate in vitro conditions identical to those that a standard, full-sized, leukocyte-depleting filter would experience during clinical cardiopulmonary CPB.

Figure 1.
Figure 1.:
Diagrammatic representation of the microcircuit used in the laboratory studies. A blood volume of 200 ml was required to prime the whole circuit. The blood was placed in the reservoir and pumped around the circuit to the heat exchanger, where its temperature could be controlled, before it was passed through the miniaturized (micro) leukocyte-depleting filter. Pressure monitors at positions 1 and 2 were used to record the prefilter and postfilter pressures.

The miniature circuit was made from polyvinyl chloride tubing, with the exception of the pump tubing, which was made of silicon. The heat exchanger was a standard low-prime blood cardioplegia heat exchanger (CSC14, Sorin Biomedica, Mirandola, Italy). Water for the heat exchanger was supplied from a standard heater/cooler unit (Cincinnati Sub-zero, Cincinnati, OH, USA). The temperature of the blood was measured using a Yellowsprings Series 400 intraluminal temperature probe (YSI, Inc., Yellowsprings, OH, USA), and recorded using a Yellowsprings temperature monitor (YSI, Inc.) at 5-minute intervals throughout each experiment. The temperature of the blood was measured at the outlet of the heat exchanger, sited approximately 1.5 cm from the entrance of the filter. The blood reservoir was a modified burette from a standard pediatric blood administration set (Baxter Health care, Thetford, UK). Blood flow through the circuit was controlled using a miniature multiflow roller pump (Watson-Marlow, York, UK).

Incorporated into the miniature circuit were specially constructed miniature filters containing the same filter media that is used for the construction of the commercially available LG6 leukocyte-depleting arterial line filter. The miniature filters were manufactured as a 4.7-cm-diameter nonpleated disc filter, with suitable connections for the microcircuit. The area of the filter was then determined and, from this, the blood volume and the flow rate requirements of the miniature circuits were calculated. The area of filter media was calculated using the formula: πr2.

To recreate in vitro the conditions that the full-size LG6 filter experiences during CPB, the scale size of the filters was calculated using the formula: Ssz = SAn / SAm, where Ssz = Scale size, SAn = Surface area of normal-sized LG6 (cm2), and SAm = Surface area of miniature filter (cm2). The scale size of the filters was then used to determine the sample volume and flow rate for the microcircuit. This made it possible to expose the test filter media to similar levels of blood elements and leukocytes, per square centimeter, that the full-size filter is exposed to during CPB. The control circuits were constructed in an identical manner, but without the incorporation of the miniature LG6 filter.

Calculation of Blood Volume and Flow for the Miniature Circuit

The area of the filter media within the miniature leukocyte-depleting (LG6) filter was calculated as follows: 3.1415926 × (4.72/2) = 17.35 cm2. Thus, the surface area of the miniature LG6 filter was 17.35 cm2, compared with 480 cm2 of the full-sized LG6 filter. The scale size of the filters was calculated as follows: Scale size = 480.00/17.35 = 27.66. As an approximation, the scale size of 1:28 (1:28 = 3.57%) was used in all subsequent calculations.

Before the start of clinical CPB, approximately 1500 ml fluid is added to prime the extracorporeal circuit leading to an increase in total circulating volume of the patient to approximately 5500 ml. Accordingly, the sample volume (or total circulating volume) required for the microcircuit was determined using the 1:28 (or 3.57%) scale size as follows: Sample volume = 5500 × 3.57% = 196.35 ml. As an approximation and in order to allow for the small volumes of blood samples drawn from the circuit at various time intervals, a sample volume of 200 ml was used in the experiments.

The standard blood flow rate during CPB in routine cardiac surgery is 5.0 l/min. By using this value as a flow rate, along with the determined scale size for the miniature circuit and filter as a reference, the blood flow required for the microcircuit was defined as follows: Circuit blood flow = 5000 ml/min × 3.57% = 178 ml/min. Accordingly, a blood flow rate of 180 ml/min was used during these experiments.

Evaluation of Blood Temperature

Throughout all experiments, the blood temperature within the circuit and the inlet and outlet pressures of the filter across the miniature filter were recorded every 5 minutes. The results of the validation revealed an error between the thermocouple 400 temperature probe and the mercury thermometer of 12% at 5°C and 0.9% at 45°C. Because the error was <1% in the 35–40°C range, and given that a temperature of 36°C was used in the experiments, this was considered acceptable.

Counting of White Cells

Total white cell counts were determined manually using the following method. Blood samples (100 μl) were diluted 10-fold by the addition of 900 μl of 0.2-μm-filtered Leucoplate solution, which causes lysis of the erythrocytes (Laboratoires Sobioda SA, Cedex, France). After thorough mixing, the suspension was allowed to stand for 10 minutes at room temperature until lysis of the red cells occurred. Then, 100 μl of this sample was carefully loaded into a Nageotte counting chamber. The chamber was left to rest in a moistened petri dish for 15 minutes to allow the leukocytes to settle. Leukocyte counts were then performed within the next 30 minutes, using light microscopy at a magnification of ×25.

The Nageotte counting chamber has a marked grid of 50 rectangles (channels). Each of these channels is 0.25 mm wide, 10 mm long, and 0.5 mm deep. A single channel contains a sample volume of 1.25 μl. During the experiments, the total number of leukocytes were counted in two channels (2.5 μl) as indicated in light gray, and recorded for subsequent analysis.

Counting of Activated White Cells

The activated white cell counts in each of the samples were determined using the Nitroblue Tetrazolium (NBT) reduction test. NBT is a water-soluble dye, which is converted to the insoluble, dark-blue Formazan by reduction. The reduced NBT is located mainly within and around the phagocytic vacuoles of activated white cells, where it acts as a visible marker of phagocytosis when the cells are examined under a light microscope. Nitroblue tetrazolium is reduced by a chemical reaction between the dye and O2, which is generated by leukocytes going through a phase of oxidative burst: NBT (yellow) + O2 = Formazan (blue) + O2. This test was performed in the following way. A stock solution of 0.4% (w/v) NBT diluted in 0.34% (w/v) sucrose, was further diluted with an equal volume of phosphate-buffered saline producing the working NBT solution. Then, 100 μl of NBT working solution was added to 100 μl of whole blood. After thorough mixing, the sample was incubated in a water bath at 37°C for 20 minutes, and left for 20 minutes at room temperature. Then, 800 μl of Leucoplate solution was added to produce a dilution of 1 in 10 of the original blood sample. After mixing, the sample was left for 10 minutes at room temperature to allow lysis of the red cells. From this solution, 100 μl was loaded into the Nageotte chamber and the number of activated cells counted. Because the number of activated cells is much smaller than the total number of leukocytes, and in order to reduce the risk of counting errors, 10 channels (125 μl) of the Nageotte chamber were counted.

Correction of WCC for the Effect of Hemodilution

Hemoglobin levels were measured in each sample to correct for the effects of hemodilution occurring during CPB, by placing the blood sample into a microcuvette, which was loaded into a HemoCue photometer that directly measures β-hemoglobin concentrations. The following formula was used for correction: [(sample hemoglobin ÷ mean hemoglobin) × white cell count]. The number of total and activated cells counted per chamber (1.25 μl) was expressed as the number of cells × 109 per liter.

Analysis of the Expression of Cd11b, CD18, and Cd62l

The analysis of the expression of the surface adhesion molecules CD11b, CD18, and CD62L was performed in a FACScalibur flow cytometer (Becton & Dickinson, UK) using fluorochrome labeled antibodies specific to each of the antigens under study.

The CD11b-PE antibody recognizes a human leukocyte antigen that is the C3bi complement receptor (CR3). CD11b is specific for the 165 kd α-subunit of the CD11b/CD18 antigen heterodimer. The CD18-PE antibody reacts with the CD18 (LFA-1) that is a transmembrane protein, β2 integrin, which associates with the α chains of CD11α, CD11b, or CD11c as a heterodimer. The CD62L-PE antibody recognizes the CD62L antigen, a leukocyte endothelial cellular adhesion molecule, which belongs to the selectin family of cell adhesion molecules.

In appropriately labeled TruCount tubes, 20 μl of the relevant antibody (anti-CD11b, anti-CD18, or anti-CD62L) and 50 μl blood were added and mixed using a vortex mixer. The tubes were incubated in the dark for at least 15 minutes. After adding 400 μl FACS erythrocyte lysing solution, the tubes were incubated for a further 10 minutes in the dark and then analyzed. The expression of surface adhesion antigens was presented as the absolute number of fluorescence events and as the percent changes in the number of these events.

Statistical Analysis

Continuous variables are expressed as the mean values and standard deviations and the proportions as percentages. The differences between the groups for the categorical variables were compared using chi-squared or Fisher exact test as appropriate. Continuous variables within the two groups with normally distributed data were compared with an unpaired student t test. If the data were skewed, a nonparametric test (Kruskal-Wallis) was used. A one-way analysis of variance (ANOVA) was used to analyze the significance of changes occurring at serial sampling intervals within each group. The overall effect of intervention was assessed by testing the sequentially obtained data in both groups with a two-way ANOVA with replicates. The degree of correlation between variables was examined with the Pearson's correlation coefficient and linear regression tests. A p value of less than or equal to 0.05 was considered statistically significant. Statistical analysis was performed on Excel 2000 (Microsoft, Redmond, WA, USA) and SPSS (SPSS, version 11.5, Chicago, IL, USA) software.


Demographics and Biochemical and Hematologic Profiles

The patients in the two groups who offered 200 ml blood each for the in vitro experiments had similar demographics and hematologic and biochemical profiles preoperatively (Table 1). Arterial blood gas, hemoglobin, and electrolyte values in the blood samples that were taken from these patients 5minutes after the start of CPB were also similar in the two groups (Table 1).

Table 1
Table 1:
Preoperative Demographic Data and Biochemical and Hematologic Profiles in the Two Groups

Reduction Rate of Manually Counted Total White Cells

In the control group, total WCCs were reduced to 90% of the baseline value (p = 0.04) over the course of the experiment, reflecting their nonspecific absorption by the tubing of the circuit. In the leukodepleted group, total WCCs fell to 51% of the baseline value (p < 0.0001). The differences between the two groups in terms of percent changes in the manually determined WCC were highly significant (p < 0.0001) (Figure 2).

Figure 2.
Figure 2.:
Reduction rate (%) of the manually determined white cell counts (WCCs) in the two groups (p < 0.0001). Error bars represent the standard deviation from the mean values.

Rate of Change in Activated White Cells

In the control group, the activated WCC rose to greater than double (116%) the baseline value after 60 minutes (p = 0.01). In the leukodepleted samples, activated leukocytes appeared to be removed in large numbers by the LG6 filter, such that numbers fell to only 14% of the baseline value over the 60 minutes of the experiment (p < 0.0001) (Figure 3). The differences in the percent changes in the activated WCCs between the two groups over time were highly significant (p < 0.0001).

Figure 3.
Figure 3.:
Changes (%) in activated white cell counts (WCCs) in the two groups (p < 0.0001). Error bars represent the standard deviation from the mean values.

Change in Cd11b Expression

In the control samples, CD11b expression increased at each sampling point, and at 60 minutes was 24% higher than the baseline (p = 0.7). In the leukodepleted samples, CD11b expression followed a different course, decreasing at each sampling point and falling to approximately 40% of the baseline values at 60 minutes (p < 0.0001) (Figure 4). The differences in the percent changes in the expression of the CD11b between the two groups over time were highly significant (p < 0.0001).

Figure 4.
Figure 4.:
Changes (%) in CD11b expression in the two groups (p < 0.0001). Error bars represent the standard deviation from the mean values.

Change in Cd18 Expression

In the control samples, CD18 expression increased by approximately 6% above baseline values after 60 minutes (p = 0.7). In the leukodepleted samples, CD18 expression was reduced at an even rate throughout the experiment and at 60 minutes was approximately by 21% less than the baseline values (p = 0.08). The differences in the percent changes in the expression of the CD18 between the two groups over time were statistically significant (p = 0.001) (Figure 5).

Figure 5.
Figure 5.:
Changes (%) in CD18 expression in the two groups (p = 0.001). Error bars represent the standard deviation from the mean values.

Change in CD62l Expression

In the control samples, CD62L expression rose by 28% above baseline after 60 minutes (p = 0.75), suggesting increasing leukocyte activation over time. In the leukodepleted samples, CD62L expression fell markedly, in a manner similar to that seen for the CD11b, and was reduced to 21% of baseline levels (p < 0.0001) (Figure 6). The differences in the expression of the CD62L over time between the two groups were highly significant (p < 0.0001).

Figure 6.
Figure 6.:
Changes (%) in CD62L expression in the two groups (p < 0.0001). Error bars represent the standard deviation from the mean values.

Correlation between Methods Used to Quantify Leukocyte Activation

A significant positive correlation was identified between the manually counted activated leukocytes and the expression of each one of the CD11b (r = 0.95, p = 0.001), CD18 (r = 0.76, p = 0.04), and CD62L (r = 0.86, p = 0.01) markers during leukodepletion.


During the circulation of fresh, whole, heparinized human blood in a simulated extracorporeal CPB circuit, there is continuous leukocyte activation as evidenced by a significant increase of the activated WCCs and a rise in the expression of the leukocyte activation markers CD11b, CD62L, and CD18. Application of a leukocyte-depletion filter to the circuit (in addition to lowering the total WCCs) leads to a dramatic reduction of activated WCCs, which is accompanied by a significant reduction in the expression of CD11b, CD62L, and CD18.

Evaluation of Filters and Leukocyte Activation in Simulated Extracorporeal Circuits

The materials used to construct the circuits and the fresh whole human blood, taken from heparinized patients connected to CPB, created conditions that resemble the contact activation phase of clinical CPB. These conditions, however, do not cover the period of ischemia and reperfusion, during which the activation status of leukocytes changes considerably. However, when assessing the efficiency of a leukodepleting device, the use of experimental extracorporeal circuits helps to avoid certain confounding factors encountered in clinical CPB, such as the margination of leukocytes in the lungs and the peripheral vasculature, the interaction of leukocytes with the reticulo-endothelial system, and the continuous release of large numbers of leukocytes from the bone marrow. In addition, in the present study, the filter media were exposed to blood at volumes comparable to that which a full-size filter would be exposed to during cardiac surgery. Therefore, use of these circuits to evaluate the quantitative and qualitative features of leukodepleting filters during the contact phase of CPB might be advantageous.

Overall Leukodepleting Performance of the LG6 Filter

The LG6 filter was found to reduce the total WCC by 49% compared with baseline values after 60 minutes in the circuit, of which 39% occurred in the first 30 minutes and 10% in the remaining 30 minutes of the experimental run. In previous studies, conducted in a similar setting, it was found that the greatest drop in the total WCC occurs during the first 20 minutes of the experimental run, suggesting that the ability of the LG6 filter to remove leukocytes is reduced upon its prolonged exposure to the circulating blood.16 To address the issue, selective application of arterial line leukodepletion during the early reperfusion phase of CPB bypass has been proposed by other investigators.25,26

Expression of Cd62l, Cd11b, and Cd18, and Activated Leukocytes in the Control Circuits

L-selectin (CD62L) is found on the surface of granulocytes, monocytes, and lymphocytes. Its expression increases after leukocyte activation, before it is rapidly shed. Thus, a reduction in L-selectin expression on the leukocyte is indicative of previous leukocyte activation. L-selectin is involved in the initial rolling of leukocytes along the endothelium mediated by loose attachment to the endothelial cell surface.27 During and after CPB, the expression of L-selectin has been seen to increase28 or to remain unchanged.29 These disparities may reflect the rapidity of changes in the expression of L-selectin and differences in the timing of the blood sampling. Despite this, L-selectin is thought to be a reliable marker of leukocyte activation.27 The increase in the expression of CD62L (L-selectin) in this study is in keeping with previous reports.27–29

Integrins consist of noncovalently bound α and β chains and are classified according to their β chain into β1 and β2 integrins. The β2-integrins (CD11/CD18 complex) are found on the surface of neutrophils, monocytes, and lymphocytes. They have a common β2 (CD18) chain and are subdivided according to their α-chain into CD11a/CD18, CD11b/CD18, and CD11c/CD18 integrins. Changes in the expression of the common β2(CD18) chain can result from the upregulation of any α-chain (a, b or c) and β2 chain combination. The expression of CD11b is promoted by several proinflammatory mediators. After the initial tethering of leukocytes to the endothelium by the selectins, firm adherence is mediated by interaction of CD11b/CD18, and its counterligand intercellular adhesion molecule-1 on the surface of the endothelium.27 The expression of β2 on neutrophils was shown to increase during CPB in a number of studies.27–29 In the control circuits, the expression of CD11b and CD18 increased over the 60-minute period of the circuit runs, albeit to a markedly different extent. As discussed earlier, CD18 is the common β chain of the β2-integrins. Changes in the expression of any of the α subunits affect the expression of the common CD18 (β2) chain. Of the three types of β2-integrin, the expression of CD11b/CD18 is known to consistently rise during and after CPB. In contrast, changes in CD11c expression are variable, whereas CD11a expression does not appear to increase during clinical and experimental CPB.27 Consequently, the lower expression of the CD18 chain, compared with CD11b/CD18, observed in this study may well reflect a modest increase, or lack of an increase, in the expression of the subunits CD11c and CD11a.

The activated WCC doubled when compared with baseline values, attesting to the leukocyte activation that takes place in the experimental runs. This rise in the activated WCC was, in fact, consistent with that observed during CPB in clinical studies previously described by our group,22,23,30 suggesting that the experimental setting used here does re-create conditions comparable to those encountered during the contact activation phase of clinical CPB.

Effect of the Leukodepleting Filter on the Expression of CD11b, CD18, and CD62L, and on the Numbers of Activated Leukocytes

The expression of CD11b and CD62L decreased significantly during the experimental runs in circuits containing a leukodepleting filter. In addition, there was an accompanying relative decrease in the expression of CD18. When compared with control circuits, the changes in the expression of each of the three surface adhesion antigens were statistically significant. A previous study using in vitro extracorporeal circulation found a 20% reduction in the surface adhesion molecules CD11b, CD11am and CD62L in the leukodepleted group, and no changes in the control group.14 In a recent report, rolipram (a phosphodiesterase type 4 inhibitor) was seen to inhibit changes in the expression of adhesion molecules and interleukin-6 release during simulated extracorporeal circulation of heparinized human blood.31,32

The activated WCCs were reduced by a total of 86% after 60 minutes of leukodepletion. This was a dramatic reduction, considering that, in the control circuits, the numbers of activated circulating leukocytes were seen to rise by 116%.

Correlation between the NBT Test and Flow Cytometric Methods of Leukocyte Activation Quantification

The NBT test, which was originally used to diagnose pyogenic infection,33 allows the identification and accurate counting of activated leukocytes in experimental and clinical settings.16,22,23,30 Flow cytometry is routinely used to identify the cellular expression of a variety of antigens. A recent study comparing the ability of the NBT test and a flow cytometric assay to quantify leukocyte activation by measuring the neutrophil respiratory burst in whole blood has found a significant correlation between the two techniques (r = 0.76 and p = 0.01).34 The present study identified a significant correlation between the reduction of the manually counted activated leukocytes and the fall in the expression of each one of the CD11b, CD18, and CD62L during leukodepletion. This shows that the NBT test and the flow cytometric analysis yield comparable and valid information on the activation status of leukocytes.

Hypothesis of Preferential Removal of Activated Leukocytes

Because activated leukocytes are more adhesive than their nonactivated counterparts,12,13 it would be reasonable to hypothesize that activated leukocytes would be removed at a higher rate through an active adhesion mechanism. Immunologic analysis of the cells trapped within the filter media that have been used in cardiac surgical procedures showed that the greater part of the trapped granulocytes were activated granulocytes.15 Others studied the effect of the metabolic inhibitor sodium azide on the leukodepletion process and observed a reduction in the rate of granulocyte removal, but not of lymphocyte removal by the filter media.17 Clinical studies on patients undergoing heart valve and coronary artery bypass graft operations documented a significantly lower expression of CD18,18 CD11b, CD62L,19 and significantly reduced activated WCCs with normothermic or hypothermic CPB following leucodepletion.22,23,30


The results of the present study are in line with those of previous reports15–19,22,23 and lend firm support to the hypothesis that leukocyte-depleting filters preferentially remove the activated forms of circulated leukocytes. These findings could explain the clinical benefits that may arise from the use of leukodepletion during clinical CPB.20,35


1. Edmunds LH Jr: Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 66: S12–S16 ; discussion S25–S38, 1998.
2. Edmunds LH Jr: The evolution of cardiopulmonary bypass: Lessons to be learned. Perfusion 17: 243–251, 2002.
3. Ohri SK: Systemic inflammatory response and the splanchnic bed in cardiopulmonary bypass. Perfusion 11: 200–212, 1996.
4. Asimakopoulos G: The inflammatory response to cardiopulmonary bypass: the role of leucocyte filtration. Perfusion 17: 7–10, 2002.
5. Morris SJ: Leucocyte reduction in cardiovascular surgery. Perfusion 16: 371–380, 2001.
6. Dzik S: Mechanisms of leucocyte removal by filtration, in Lane TA, Myllyla G (eds), Leucocyte Depleted Blood Products. Karger, Basel, 1994, pp 123.
7. Steneker I, van Luyn MJA, van Wachem PB, Biewenga J: Electron microscopic examination of white cells depletion on four leucocyte depletion filters. Transfusion 32: 450–457, 1992.
8. Steneker I, Prins HK, Florie M, et al.: Mechanisms of white cell reduction in red cell concentrates by filtration: The effect of the cellular composition on the red cell concentrates. Transfusion 33: 42–50, 1993.
9. Gu YJ, de Vries AJ, Boonstra PW, van Oeveren WV: Mechanism and strategy of leucocyte depletion during cardiopulmonary bypass surgery, in Matheis G, Moritz A, Scholz M (eds), Leucocyte Depletion in Cardiac Surgery and Cardiology. Karger, Basel, 2002, pp 78.
10. van Oeveren W. Leucocyte and platelet activation during extracorporeal circulation. Cell Mater 4: 187–190, 1994.
11. Gu YJ, van Oeveren W: Activation of plasma components by leucocyte removal filters. ASAIO J 40: M598–M601, 1994.
12. Berger M, O'Shea J, Cross AS, et al.: Human neutrophils increase expression of Crbi as well as C3b receptors upon activation. J Clin Invest 74: 1566–1571, 1984.
13. Zimmerman GA, Renzetti AD, Hill AR: Functional and metabolic activity of granulocytes from patients with adult respiratory distress syndrome: Evidence for activated neutrophils in the pulmonary circulation. Am Rev Respir Dis 127: 290–300, 1983.
14. Thurlow PJ, Doolan L, Sharp R, et al.: Laboratory studies of the effect of Pall extracorporeal leucocyte filters LG6 and AV6 on patients undergoing coronary bypass grafts. Perfusion 11: 29–37, 1996.
15. Smit JJ, de Vries AJ, Gu YJ, van Oeveren W: Filtration of activated granulocytes during cardiopulmonary bypass: A morphologic and immunologic study to characterize trapped leucocytes. J Lab Clin Med 135: 238–246, 2000.
16. Sheppard SV: Leucocyte filtration and cardiac surgery [PhD thesis]. Portsmouth, UK, University of Portsmouth, 1999.
17. Oka S, Maeda K, Nishimura T, Yamawaki N: Mechanisms of leukocyte removal with fibers, in Sekiguchi S (ed), Clinical Application of Leukocyte Depletion. Oxford, UK, Blackwell Scientific, 1993, pp105.
18. Hurst T, Johnson D, Cujec B, et al.: Depletion of activated neutrophils by a filter during cardiac valve surgery. Can J Anaesth 44: 131–139, 1997.
19. Chen YY, Tsai WC, Lin CC, et al.: Leucocyte depletion attenuates expression of neutrophil adhesion molecules during cardiopulmonary bypass in human beings. J Thorac Cardiovasc Surg 123: 218–224, 2002.
20. Stefanou D, Gourlay T, Asimakopoulos G, Taylor KM: Leucodepletion during cardiopulmonary bypass reduces blood transfusion and crystalloid requirements. Perfusion 16: 51–58, 2001.
21. Johnson D, Thomson D, Mycyk T, et al.: Depletion of neutrophils by filtration during aortocoronary bypass surgery transiently improves postoperative cardiorespiratory status. Chest 107: 1253–1259, 1995.
22. Alexiou C: Leucocyte activation and effects of leucocyte depletion during clinical and experimental extracorporeal circulation [PhD thesis]. Portsmouth, UK, University of Portsmouth 2004.
23. Alexiou C, Sheppard SV, Smith DC, et al.: The effect of blood temperature on the efficacy of systemic leucodepletion during cardiopulmonary bypass: A prospective randomized clinical study. ASAIO J 51: 802–807, 2005.
24. Edmunds LH Jr: Pathologic physiology of cardiopulmonary bypass, in: Edmunds LH Jr (ed), Cardiac Surgery in the Adult. New York, McGraw-Hill, 1997.
    25. Matheis G, Scholz M, Gerber J, et al.: Leucocyte depletion in the early reperfusion phase on cardiopulmonary bypass reduces myocardial injury. Perfusion 16: 43–49, 2001.
    26. Matheis G, Simon A, Scholz M: Strategic leucocyte filtration: Clinical and experimental experience, in Matheis G, Moritz A, Scholz M (eds), Leucocyte Depletion in Cardiac Surgery and Cardiology. Karger, Basel, 2002, pp 101.
    27. Asimakopoulos G, Taylor KM: Effects of cardiopulmonary bypass on leucocyte and endothelial adhesion molecules. Ann Thorac Surg 66: 2135–2144, 1998.
    28. Finn A, Rebuck N, Moat N: Leucocyte activation during cardiopulmonary bypass. J Thorac Cardiovasc Surg 104: 1746–8, 1992.
    29. Galinanes M, Watson C, Trivedi U, et al.: Differential patterns of neutrophil adhesion molecules during cardiopulmonary bypass in humans. Circulation 96: 364–369, 1996.
    30. Alexiou C, Tang AT, Sheppard SV, et al.: The effect of leucodepletion on leucocyte activation, pulmonary inflammation and respiratory index in surgery for coronary revascularisation: A prospective randomized study. Eur J Cardiothorac Surg 26: 294–300, 2004.
    31. Sato Y, Hiramatsu Y, Homma S, et al.: Phosphodiesterase type 4 inhibitor rolipram inhibits activation of monocytes during extracorporeal circulation. J Thorac Cardiovasc Surg 130: 346–350, 2005.
    32. Sato Y, Hiramatsu Y, Homma S, et al.: Phosphodiesterase type 4 inhibition of activated polymorphonuclear leukocytes in a simulated extracorporeal circulation model. J Thorac Cardiovasc Surg 125: 172–177, 2003.
    33. Park BH, Fikrig SM, Smithwick SM: Infection and nitroblue-tetrazolium reduction by the neutrophils. Lancet 2: 532–534, 1968.
    34. Richardson MP, Ayliffe MJ, Helbert M, Davies EG: A simple flow cytometric assay using dihydrorhodamine for the measurement of the neutrophil respiratory burst in whole blood: Comparison with the quantitative nitrobluetetrazolium test. J Immunol Methods 219: 187–193, 1998.
    35. Tang ATM, Alexiou C, Hsu J, et al.: Leucodepletion during cardiopulmonary bypass reduces renal injury in coronary revascularization: A prospective randomized study. Ann Thorac Surg 74: 372–377, 2002.
    Copyright © 2006 by the American Society for Artificial Internal Organs